The role of valve surgery in infective endocarditis management: A systematic review of observational studies that included propensity score analysis

Abstract

The potential role of valve surgery in infective endocarditis (IE) management is controversial. No randomized trials have been conducted to date; accordingly, some studies use propensity score analysis (PSA) to minimize selection bias in observational studies. A systematic review of the literature addressing the role of valve surgery in IE was performed. Studies in which PSA was applied to the management of IE were identified using Medline, Web of Science, Zetoc, and Article First from inception to June 2007. Cohort studies that compared valve surgery (combined with antimicrobial therapy) to antimicrobial therapy alone and used PSA to adjust for selection bias were eligible. Methodological details and outcomes were compared to assess methodological quality. Six cohort studies that enrolled a total of 3,409 patients with either native valve endocarditis (NVE) or prosthetic valve endocarditis (PVE) were eligible. The proportions of IE cases undergoing valve surgery ranged from 28.9% to 41.7% for PVE and from 20.8% to 49.3% for NVE. Two studies suggested a statistically significant survival benefit with surgery in patients with NVE. Data from 2 other patient cohorts (PVE and NVE) revealed conflicting results. No statistically significant associations were found in the remaining studies. There are several limitations of available cohort studies that assess the role of valve surgery in IE. Well-designed prospective studies that address these limitations are needed to further define the role of surgery in IE. Until then, careful scrutiny is warranted when making management decisions in complicated, left-sided IE.